class Prawn::Images::JPG
A convenience class that wraps the logic for extracting the parts of a JPG image that we need to embed them in a PDF
Constants
- JPEG_SOF_BLOCKS
Attributes
bits[R]
@group Extension API
channels[R]
@group Extension API
height[R]
@group Extension API
scaled_height[RW]
scaled_width[RW]
width[R]
@group Extension API
Public Class Methods
can_render?(image_blob)
click to toggle source
# File lib/prawn/images/jpg.rb, line 24 def self.can_render?(image_blob) image_blob[0, 3].unpack("C*") == [255, 216, 255] end
new(data)
click to toggle source
Process a new JPG image
:data
-
A binary string of JPEG data
# File lib/prawn/images/jpg.rb, line 32 def initialize(data) @data = data d = StringIO.new(@data) d.binmode c_marker = 0xff # Section marker. d.seek(2) # Skip the first two bytes of JPEG identifier. loop do marker, code, length = d.read(4).unpack('CCn') fail "JPEG marker not found!" if marker != c_marker if JPEG_SOF_BLOCKS.include?(code) @bits, @height, @width, @channels = d.read(6).unpack("CnnC") break end d.seek(length - 2, IO::SEEK_CUR) end end
Public Instance Methods
build_pdf_object(document)
click to toggle source
Build a PDF object representing this image in
document
, and return a Reference to it.
# File lib/prawn/images/jpg.rb, line 55 def build_pdf_object(document) color_space = case channels when 1 :DeviceGray when 3 :DeviceRGB when 4 :DeviceCMYK else fail ArgumentError, 'JPG uses an unsupported number of channels' end obj = document.ref!( :Type => :XObject, :Subtype => :Image, :ColorSpace => color_space, :BitsPerComponent => bits, :Width => width, :Height => height ) # add extra decode params for CMYK images. By swapping the # min and max values from the default, we invert the colours. See # section 4.8.4 of the spec. if color_space == :DeviceCMYK obj.data[:Decode] = [ 1.0, 0.0, 1.0, 0.0, 1.0, 0.0, 1.0, 0.0 ] end obj.stream << @data obj.stream.filters << :DCTDecode obj end